Bipolar I disorder (BPI) is a severe psychiatric disorder with a strong genetic influence on susceptibility. Heritability, the proportion of phenotypic variation in a population that is attributable to genetic variation, is estimated to be >90%. However, intense efforts using both linkage and association studies to identify susceptibility loci thus far have met only limited success. Recently it has been appreciated that widespread copy number variation (CNV), in the form of duplications and deletions, frequently occurs in the human genome and may be a major contributor to individual genetic variation. However, CNV has not been sampled in typical genetic studies, and may therefore be an unrecognized source of BPI genetic susceptibility and potentially a confounding influence on prior investigations of disease. We propose here to characterize CNV in 366 case-parent BPI trios plus an additional 162 BPI cases, all of Ashkenazi Jewish descent. We will survey the entire nonrepetitive human genome, at an average density of 1.1 kb, using 2.1 million feature oligonucleotide arrays and a competitive genomic hybridization protocol, reliably detecting CNV 8kb and larger. For common (>1% frequency) CNV, the data will be evaluated for distorted transmission of CNV to affected BPI offspring. Additionally, presence of excess rare (<1%) CNV in BPI linkage regions as well as in or near previously identified candidate gene regions will be evaluated by comparison of 510 independent BPI cases and 500 Ashkenazi controls previously assessed for genome-wide CNV. Significant CNV loci will be validated with an alternate technology, breakpoints will be resolved with PCR and/or quantitative genotyping strategies, and variants will be carefully scrutinized for their proximity to genes or evolutionarily conserved sequences. We further propose to explore association in a joint dataset of the 366 BPI trios and 162 BPI cases discussed here, plus 500 schizophrenia (SZ) patients (including 300 SZ trios) and 500 controls, all of Ashkenazi Jewish descent, in whom CNV detection is already proceeding. It has been suggested that BPI and schizophrenia may not represent distinct entities, but instead are varying manifestations of a similar biological lesion. This is supported by the substantial overlap between these disorders, in both clinical presentation and genomic loci identified by linkage and association. Our joint dataset, the largest of its kind from an inbred population, has the potential to test the assertion that BPI and SZ are part of a continuum of psychiatric illness, and reveal CNV that predispose to psychiatric illness. PUBLIC HEALTH RELEVANCE: Bipolar I disorder is a severe psychiatric illness that affects ~1% of the general population, but causes of this disorder remain unknown. We propose to investigate whether deletions or duplications in the human genome are related to bipolar I susceptibility;these variants may harbor important clues about the genes, and ultimately the biological process, involved in the development of Bipolar I disorder.